表格问答研究综述

doi:10.11896/jsjkx.250900006

摘要/Abstract

摘要： 表格作为一种重要的数据载体,能以紧凑的形式承载大量高价值信息,被广泛应用于经济、金融及科研等领域。表格问答(Table Question Answering,TableQA)旨在针对用自然语言描述的问题,从表格数据中自动进行推理并生成相应的答案,是自然语言处理与数据分析交叉领域的重要研究方向。与传统的文本问答和知识库问答相比,表格问答不仅需要理解自然语言,还须解析表格的二维结构,并处理数值计算与复杂逻辑推理,因此面临更大的挑战。近年来,随着多样化数据集的持续构建,表格问答技术不断取得进展。其研究范式经历了从基于规则与模板的方法,到统计学习与神经网络模型的应用,再到预训练语言模型的引入,整体性能不断提升。尤其是近年来大语言模型(Large Language Models,LLMs)的兴起,进一步推动了表格问答进入新的发展阶段。凭借卓越的跨任务泛化能力与推理能力,大语言模型加速了新型研究范式的形成与发展,为方法创新提供了有力支撑。文中系统梳理了表格问答技术的演进脉络与代表性方法,重点总结了大语言模型驱动下的最新研究进展,概述了当前研究面临的关键挑战,并对未来发展趋势进行了展望。

关键词: 表格问答, 表格推理, 大语言模型, 自然语言处理

Abstract: Tables serve as essential data carriers,capable of efficiently storing large volumes of high-value information.They are widely used across domains such as economics,finance,scientific research,and more.Table question answering(TableQA) aims to automatically derive answers from tabular data in response to natural language queries,representing a key research direction at the intersection of natural language processing and data analysis.Compared to traditional text-based or knowledge-base question answering,TableQA presents greater challenges,as it requires not only natural language understanding but also the interpretation of two-dimensional table structures,numerical computations,and complex logical reasoning.In recent years,the continuous deve-lopment of diverse datasets has driven steady progress in TableQA research.The field has evolved from early rule-based and template-based approaches to statistical learning and neural network models,and more recently,to the integration of pre-trained language models,resulting in consistent performance improvements.Notably,the emergence of large language models(LLMs) has ushered in a new phase of development.Leveraging their strong cross-task generalization and reasoning capabilities,LLMs have accelerated innovation and fostered new research paradigms in TableQA.This paper systematically reviews the evolution and representative methods of TableQA,with a particular emphasis on recent advances enabled by LLMs.It also outlines the key challenges currently facing the field and provides a forward-looking perspective on future research directions.

Key words: Table question answering, Table reasoning, Large language model, Natural language processing

中图分类号:

TP391

吴贤杰, 李彤亮, 李舟军. 表格问答研究综述[J]. 计算机科学, 2026, 53(3): 295-306. https://doi.org/10.11896/jsjkx.250900006

WU Xianjie, LI Tongliang, LI Zhoujun. Survey of Table Question Answering Research[J]. Computer Science, 2026, 53(3): 295-306. https://doi.org/10.11896/jsjkx.250900006

参考文献

[1]LU W,ZHANG J,FAN J,et al.Large language model for table processing:A survey[J].Frontiers of Computer Science,2025,19(2):192350.
[2]DONG H,ZHAO J,TIAN Y,et al.SpreadsheetLLM:encoding spreadsheets for large language models[J].arXiv:2407.09025,2024.
[3]NAVEED H,KHAN A U,QIU S,et al.A comprehensive overview of large language models[J].ACM Transactions on Intelligent Systems and Technology,2025,16(5):1-72.
[4]LI P,HE Y,YASHAR D,et al.Table-gpt:Table fine-tuned gpt for diverse table tasks[J].Proceedings of the ACM on Management of Data,2024,2(3):1-28.
[5]BADARO G,SAEED M,PAPOTTI P.Transformers for tabular data representation:A survey of models and applications[J].Transactions of the Association for Computational Linguistics,2023,11:227-249.
[6]CHENG Z,DONG H,WANG Z,et al.HiTab:A HierarchicalTable Dataset for Question Answering and Natural Language Generation[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2022:1094-1110.
[7]DONG H,CHENG Z,HE X,et al.Table Pre-training:A Survey on Model Architectures,Pre-training Objectives,and Downstream Tasks[C]//Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence.2022.
[8]GU Y,PAHUJA V,CHENG G,et al.Knowledge Base Question Answering:A Semantic Parsing Perspective[C]//AKBC.2022.
[9]NASSIRI K,AKHLOUFI M.Transformer models used for text-based question answering systems[J].Applied Intelligence,2023,53(9):10602-10635.
[10]LIU X,SHEN S,LI B,et al.A Survey of NL2SQL with Large Language Models:Where are we,and where are we going?[J].arXiv:2408.05109,2024.
[11]KATSIS Y,CHEMMENGATH S,KUMAR V,et al.AIT-QA:Question Answering Dataset over Complex Tables in the Airline Industry[C]//Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies:Industry Track.2022:305-314.
[12]SHIGAROV A.Table understanding:Problem overview[J].Wiley Interdisciplinary Reviews:Data Mining and Knowledge Discovery,2023,13(1):e1482
[13]WU X,YANG J,CHAI L,et al.Tablebench:A comprehensive and complex benchmark for table question answering[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2025:25497-25506.
[14]CHEN Z,CHEN W,SMILEY C,et al.FinQA:A Dataset of Numerical Reasoning over Financial Data[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing.2021:3697-3711.
[15]PASUPAT P,LIANG P.Compositional Semantic Parsing onSemi-Structured Tables[C]//Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing(Volume 1:Long Papers).2015:1470-1480.
[16]IYYER M,YIH W,CHANG M W.Search-based neural structured learning for sequential question answering[C]//Procee-dings of the 55th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2017:1821-1831.
[17]CHEN W,WANG H,CHEN J,et al.Tabfact:A large-scaledataset for table-based fact verification[J].arXiv:1909.02164,2019.
[18]ZHONG V,XIONG C,SOCHER R.Seq2sql:Generating structured queries from natural language using reinforcement lear-ning[J].arXiv:1709.00103,2017.
[19]YU T,ZHANG R,YANG K,et al.Spider:A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task[C]//Proceedings of the 2018 Conference on Empirical Methods in Natural Language Proces-sing.2018:3911-3921.
[20]LI J,HUI B,QU G,et al.Can llm already serve as a database interface? a big bench for large-scale database grounded text-to-sqls[J].Advances in Neural Information Processing Systems,2023,36:42330-42357.
[21]MIN Q,SHI Y,ZHANG Y.A Pilot Study for Chinese SQL Semantic Parsing[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing(EMNLP-IJCNLP).2019:3652-3658.
[22]PARIKH A,WANG X,GEHRMANN S,et al.ToTTo:A Controlled Table-To-Text Generation Dataset[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing(EMNLP).2020:1173-1186.
[23]CHEN W,CHANG M W,SCHLINGER E,et al.Open question answering over tables and text[J].arXiv:2010.10439,2020.
[24]NAN L,HSIEH C,MAO Z,et al.FeTaQA:Free-form tablequestion answering[J].Transactions of the Association for Computational Linguistics,2022,10:35-49.
[25]KRUMDICK M,KONCEL-KEDZIORSKI R,LAI V D,et al.Bizbench:A quantitative reasoning benchmark for business and finance[C]//Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2024:8309-8332.
[26]CHEN W,ZHA H,CHEN Z,et al.HybridQA:A Dataset of Multi-Hop Question Answering over Tabular and Textual Data[C]//Findings of the Association for Computational Linguistics:EMNLP 2020.2020:1026-1036.
[27]ZHU F,LEI W,HUANG Y,et al.TAT-QA:A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance[C]//Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing(Volume 1:Long Papers).2021:3277-3287.
[28]ZHENG M,HAO Y,JIANG W,et al.IM-TQA:A Chinese table question answering dataset with implicit and multi-type table structures[C]//Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2023:5074-5094.
[29]ZHAO Y,LI Y,LI C,et al.MultiHiertt:Numerical Reasoningover Multi Hierarchical Tabular and Textual Data[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2022:6588-6600.
[30]ZHOU W,MESGAR M,ADEL H,et al.FREB-TQA:A Fine-Grained Robustness Evaluation Benchmark for Table Question Answering[C]//Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies(Volume 1:Long Papers).2024:2479-2497.
[31]ASHURY-TAHAN S,MAI Y,GERA A,et al.The mightytorr:A benchmark for table reasoning and robustness[J].arXiv:2502.19412,2025.
[32]KIM Y,YIM M,SONG K Y.Tablevqa-bench:A visual question answering benchmark on multiple table domains[J].arXiv:2404.19205,2024.
[33]ZHAO W,FENG H,LIU Q,et al.Tabpedia:Towards comprehensive visual table understanding with concept synergy[J].Advances in Neural Information Processing Systems,2024,37:7185-7212.
[34]ZHENG X,BURDICK D,POPA L,et al.Global table extractor(gte):A framework for joint table identification and cell structure recognition using visual context[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision.2021:697-706.
[35]SMOCK B,PESALA R,ABRAHAM R.PubTables-1M:To-wards comprehensive table extraction from unstructured documents[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2022:4634-4642.
[36]ZHENG M,FENG X,SI Q,et al.Multimodal Table Under-standing[C]//Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2024:9102-9124.
[37]WU P,YANG Y,ZHU G,et al.RealHiTBench:A Comprehensive Realistic Hierarchical Table Benchmark for Evaluating LLM-Based Table Analysis[J].arXiv:2506.13405,2025.
[38]SINGH A,BIEMANN C,STRICH J.MTabVQA:EvaluatingMulti-Tabular Reasoning of Language Models in Visual Space[J].arXiv:2506.11684,2025.
[39]PAPINENI K,ROUKOS S,WARD T,et al.Bleu:a method for automatic evaluation of machine translation[C]//Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics.2002:311-318.
[40]CHIN-YEW L.Rouge:A package for automatic evaluation ofsummaries[C]//Proceedings of the Workshop on Text Summarization Branches Out,2004.2004.
[41]ZHANG T,KISHORE V,WU F,et al.BERTScore:Evaluating Text Generation with BERT[C]//International Conference on Learning Representations.2020.
[42]COPESTAKE A,JONES K S.Natural language interfaces todatabases[J].The Knowledge Engineering Review,1990,5(4):225-249.
[43]WOODS W A,KAPLAN R M,NASH-WEBBER B.The lunar sciences natural language information system:Final report [R].Cambridge,MA:Bolt,Beranek and Newman,Inc.,1972.
[44] ANDROUTSOPOULOS I,RITCHIE G,THANISCH P.MASQUE/SQL:an efficient and portable natural language query interface for relational databases [C]//Proceedings of the 6th International Conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems(IEA/AIE’93).1993:327-330
[45]POPESCU A M,ARMANASU A,ETZIONI O,et al.Precise on atis:semantic tractability and experimental results[C]//AAAI.2004:1026-1027.
[46]PASUPAT P,LIANG P.Inferring Logical Forms From Denotations[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2016:23-32.
[47]XU X,LIU C,SONG D.Sqlnet:Generating structured queries from natural language without reinforcement learning[J].ar-Xiv:1711.04436,2017.
[48]ZHANG S,BALOG K.Ad hoc table retrieval using semanticsimilarity[C]//Proceedings of the 2018 World Wide Web Conference.2018:1553-1562.
[49]JAUHAR S K,TURNEY P,HOVY E.Tables as semi-structured knowledge for question answering[C]//Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2016:474-483.
[50]SUN H,MA H,HE X,et al.Table cell search for question answering[C]//Proceedings of the 25th International Conference on World Wide Web.2016:771-782.
[51]ZHANG S,TUAN L A,ZHAO C.SynTQA:Synergistic Table-based Question Answering via Mixture of Text-to-SQL and E2E TQA[C]//Findings of the Association for Computational Linguistics:EMNLP 2024.2024:2352-2364.
[52]ZHONG V,XIONG C,SOCHER R.Seq2sql:Generating structured queries from natural language using reinforcement learning[J].arXiv:1709.00103,2017.
[53]YIN P,NEUBIG G,YIH W,et al.TaBERT:Pretraining forJoint Understanding of Textual and Tabular Data[C]//Procee-dings of the 58th Annual Meeting of the Association for Computational Linguistics.2020:8413-8426.
[54]HERZIG J,NOWAK P K,MÜLLER T,et al.TaPas:Weakly Supervised Table Parsing via Pre-training[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics.Association for Computational Linguistics,2020.
[55]DENG X,SUN H,LEES A,et al.Turl:Table understandingthrough representation learning[J].ACM SIGMOD Record,2022,51(1):33-40.
[56]WANG Z,DONG H,JIA R,et al.Tuta:Tree-based transfor-mers for generally structured table pre-training[C]//Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery &Data Mining.2021:1780-1790.
[57]IIDA H,THAI D,MANJUNATHA V,et al.TABBIE:Pre-trained Representations of Tabular Data[C]//Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies.2021:3446-3456.
[58]YU T,WU C S,LIN X V,et al.GraPPa:Grammar-Augmented Pre-Training for Table Semantic Parsing[C]//International Conference on Learning Representations.2021.
[59]LIU Q,CHEN B,GUO J,et al.TAPEX:Table Pre-training via Learning a Neural SQL Executor[C]//International Conference on Learning Representations.2022.
[60]BROWN T,MANN B,RYDER N,et al.Language models arefew-shot learners[J].Advances in Neural Information Proces-sing Systems,2020,33:1877-1901.
[61]DONG Q,LI L,DAI D,et al.A Survey on In-context Learning[C]//Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing.2024:1107-1128.
[62]WANG Y Z,LI Q,DAI Z J,et al.Research status and trends of large language models [J].Journal of Engineering Science,2024,46(8):1411-1425.
[63]SAHOO P,SINGH A K,SAHA S,et al.A systematic survey of prompt engineering in large language models:Techniques and applications[J].arXiv:2402.07927,2024.
[64]CHEN W.Large Language Models are few(1)-shot Table Reasoners[C]//Findings of the Association for Computational Linguistics:EACL 2023.2023:1120-1130.
[65]SINGHA A,CAMBRONERO J,GULWANI S,et al.TabularRepresentation,Noisy Operators,and Impacts on Table Structure Understanding Tasks in LLMs[C]//NeurIPS 2023 Second Table Representation Learning Workshop.2023.
[66]GAO D,WANG H,LI Y,et al.Text-to-SQL Empowered by Large Language Models:A Benchmark Evaluation[J].Proceedings of the VLDB Endowment,2024,17(5):1132-1145.
[67]CHEN W,MA X,WANG X,et al.Program of thoughts prompting:Disentangling computation from reasoning for numerical reasoning tasks[J].arXiv:2211.12588,2022.
[68]YAO S,ZHAO J,YU D,et al.React:Synergizing reasoning and acting in language models[C]//International Conference on Learning Representations(ICLR).2023.
[69]ZHENG M,YANG H,JIANG W,et al.Chain-of-thought rea-soning in tabular language models[C]//Findings of the Association for Computational Linguistics:EMNLP 2023.2023:11006-11019.
[70]WANG Z,ZHANG H,LI C L,et al.Chain-of-Table:Evolving Tables in the Reasoning Chain for Table Understanding[C]//ICLR.2024.
[71]WEI J,WANG X,SCHUURMANS D,et al.Chain-of-thoughtprompting elicits reasoning in large language models[J].Advances in Neural Information Processing Systems,2022,35:24824-24837.
[72]LI Z,WANG Z,YANG F G,et al.Research and prospects of table-based automatic question answering [J].Journal of Computer Engineering and Applications,2021,57(13).
[73]JINZ Q,LU W.Tab-CoT:Zero-shot Tabular Chain of Thought[C]//Findings of the Association for Computational Linguistics:ACL 2023.2023:10259-10277.
[74]ZHANG Y,HENKEL J,FLORATOU A,et al.Reactable:Enhancing react for table question answering[J].arXiv:2310.00815,2023.
[75]CHENG Z,XIE T,SHI P,et al.Binding language models insymbolic languages[J].arXiv:2210.02875,2022.
[76]YE Y,HUI B,YANG M,et al.Large language models are versatile decomposers:Decomposing evidence and questions for table-based reasoning[C]//Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval.2023:174-184.
[77]NGUYEN G,BRUGERE I,SHARMA S,et al.Interpretable LLM-based Table Question Answering[J].arXiv:2412.12386v3,2024.
[78]ZHAO Y,CHEN L,COHAN A,et al.TaPERA:Enhancing faithfulness and interpretability in long-form TableQA by content planning and execution-based reasoning[C]//Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers).2024:12824-12840.
[79]WANG X,WEI J,SCHUURMANS D,et al.Self-consistencyimproves chain of thought reasoning in language models[J].arXiv:2203.11171,2022.
[80]LUO H,SHEN Y,DENG Y.Unifying text,tables,and images for multimodal question answering[C].Findings of the Association for Computational Linguistics:EMNLP 2023.Association for Computational Linguistics,2023.
[81]KIM G,HONG T,YIM M,et al.Ocr-free document understanding transformer[C]//European Conference on Computer Vision.Cham:Springer Nature Switzerland,2022:498-517.
[82]LEE K,JOSHI M,TURC I R,et al.Pix2struct:Screenshot parsing as pretraining for visual language understanding[C]//International Conference on Machine Learning.PMLR,2023:18893-18912.
[83]WEI X,ZHANG T,LI Y,et al.Multi-modality cross attention network for image and sentence matching[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:10941-10950.
[84]ZHANG M,CHANG K,WU Y.Multi-modal Semantic Understanding with Contrastive Cross-modal Feature Alignment[C]//Proceedings of the 2024 Joint International Conference on Computational Linguistics,Language Resources and Evaluation(LREC-COLING 2024).2024:11934-11943.
[85]RADFORD A,KIM J W,HALLACY C,et al.Learning transferable visual models from natural language supervision[C]//International Conference on Machine Learning.PmLR,2021:8748-8763.
[86]LIU H,LI C,WU Q,et al.Visual instruction tuning[J].Advances in Neural Information Processing Systems,2023,36:34892-34916.
[87]ZHANG Z,LIU D,LIU S,et al.Turbo your multi-modal classification with contrastive learning[J].arXiv:2409.09282,2024.
[88]LIU Z,WANG H,LI X,et al.Hippo:Enhancing the table understanding capability of large language models through hybrid-modal preference optimization[J].arXiv:2502.17315,2025.
[89]ZHOU B,GAO Z,WANG Z,et al.SynTab-LLaVA:Enhancing Multimodal Table Understanding with Decoupled Synthesis[C]//Proceedings of the Computer Vision and Pattern Recognition Conference.2025:24796-24806.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed